Checkpoints for Dilgren & Wiegreffe (2026), arxiv.org/abs/2604.04902: No-CoT/CoT/Coconut/CODI + multi-mode, GPT-2 & Llama-3.2-1B.
Connor Dilgren
connordilgren
·
AI & ML interests
None yet
Recent Activity
updated a collection about 13 hours ago
Are Latent Reasoning Models Easily Interpretable? updated a model about 13 hours ago
connordilgren/llama32-1b-prosqa-multimode-codi published a model about 13 hours ago
connordilgren/llama32-1b-prosqa-multimode-codiOrganizations
None yet
models 37
connordilgren/llama32-1b-prosqa-multimode-codi
Updated
connordilgren/llama32-1b-prosqa-multimode-coconut
Updated
connordilgren/llama32-1b-prosqa-codi
Updated
connordilgren/llama32-1b-prosqa-coconut
Updated
connordilgren/llama32-1b-prosqa-cot
Updated
connordilgren/llama32-1b-prosqa-no-cot
Updated
connordilgren/llama32-1b-prontoqa-multimode-codi
Updated
connordilgren/llama32-1b-prontoqa-multimode-coconut
Updated
connordilgren/llama32-1b-prontoqa-codi
Updated
connordilgren/llama32-1b-prontoqa-coconut
Updated
datasets 0
None public yet